2020/10/28 by Alejandro González Morales, Tom Froese, Morales, Alejandro +1
Biochemistry, Genetics and Molecular Biology · #FOS: Biological sciences #FOS: Computer and information sciences #Genetics, Aging, and Longevity in Model Organisms #Machine Learning (cs.LG) #Neurons and Cognition (q-bio.NC)
paper · pdf · doi:10.48550/arxiv.2010.15272
openalex publication_date 2020/10/28 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
The nervous system of the nematode soil worm Caenorhabditis elegans exhibits\nremarkable complexity despite the worm's small size. A general challenge is to\nbetter understand the relationship between neural organization and neural\nactivity at the system level, including the functional roles of inhibitory\nconnections. Here we implemented an abstract simulation model of the C. elegans\nconnectome that approximates the neurotransmitter identity of each neuron, and\nwe explored the functional role of these physiological differences for neural\nactivity. In particular, we created a Hopfield neural network in which all of\nthe worm's neurons characterized by inhibitory neurotransmitters are assigned\ninhibitory outgoing connections. Then, we created a control condition in which\nthe same number of inhibitory connections are arbitrarily distributed across\nthe network. A comparison of these two conditions revealed that the biological\ndistribution of inhibitory connections facilitates the self-optimization of\ncoordinated neural activity compared with an arbitrary distribution of\ninhibitory connections.\n